Extrinsic Dexterity Through Active Slip Control Using Deep Predictive Models

Simon Stepputtis, Yezhou Yang, Heni Ben Amor

IEEE International Conference on Robotics and Automation (ICRA), 2018

Conference Paper

Full Paper Copy Bibtex

Content

We present a machine learning methodology for actively controlling slip, in order to increase robot dexterity. Leveraging recent insights in deep learning, we propose a Deep Predictive Model that uses tactile sensor information to reason about slip and its future influence on the manipulated object. The obtained information is then used to precisely manipulate objects within a robot end-effector using external perturbations imposed by gravity or acceleration. We show in a set of experiments that this approach can be used to increase a robot’s repertoire of motor skills.

Citation

@inproceedings{Stepputtis2018,
  doi = {10.1109/icra.2018.8461055},
  url = {https://doi.org/10.1109/icra.2018.8461055},
  year = {2018},
  month = may,
  publisher = {{IEEE}},
  author = {Simon Stepputtis and Yezhou Yang and Heni Ben Amor},
  title = {Extrinsic Dexterity Through Active Slip Control Using Deep Predictive Models},
  booktitle = {2018 {IEEE} International Conference on Robotics and Automation ({ICRA})}
}